Understanding how and why social inequalities in longevity exist is a major public health priority, requiring integrative theory testing with a view toward translation into useful clinical tools. This application emerges directly from New and Early Stage Investigator Ben Chapman's K08 research program, and is responsive to NIA priorities (PA-11-124, RFA-AG-11-004, PA-09-216, NIA Workgroup on Personality and Healthy Aging) and to calls for action issued by the Institute of Medicine (Committee Report on Genes, Behavior, and the Social Environment) and Cochrane Diagnosis Group (Statement on 10 Steps to Improving Prognostic Models). We have assembled a consortium of 18 studies spanning roughly 11,000 midlife and older adults, called Project Peril (Personality Epidemiologic Research on Inequalities in Longevity). Studies will be harmonized and linked to the National Death Index (NDI), providing follow- up data over a Mean(SD) = 12(4) year period. In Aim 1, we will use Project PERIL data to test hypotheses derived from the Social Structure and Personality (SSP) model of longevity. The SSP model proposes that the social structure contributes and individual socio-emotional and behavioral dispositions captured by Big 5 personality phenotype are interwoven. Dispositional tendencies that play an adaptive role in negotiating socioeconomic disadvantage may also compromise health over the lifespan, leading to earlier mortality. Thus, social and personality determinants of longevity may be coupled to an unknown degree. Understanding this sociostructural-psychological interface can inform public health programs and policy aimed at reducing social inequalities. In Aim 2, we test the hypothesis of cumulative disadvantage, which suggests that personality-SES mortality risk not only increase with age, but become increasingly coupled with age. As well, potential mechanisms of these risks are studied. Aim 3 leverages Aim 1 and 2 findings to develop clinical prognostic models--valid, pragmatic, and actionable prediction models of mortality risk. Prognostic models have grown in popularity with the advent of personalized medicine, and yield individualized, data- driven risk estimates for health outcomes based on a person's specific risk factor profile. Thus far, prognostic models have incorporated strictly biomedical risk factors, despite the actuarial predictive power of SES and personality. Aim 3 therefore focuses on putting the person in personalized medicine by developing improved prognostic models for 10-year mortality risk using data on SES and personality phenotype, both alone and in conjunction with standard chronic disease components of the Charlson Comorbidity Index, a popular prognostic model for mortality. Integrative Data Analysis (IDA) will be used to harness the data consortium's power and breadth, via multilevel structural equation models (Aims 1 and 2) and parametric survival models (Aim 3). The Project PERIL team includes experts in social epidemiology, personality and longevity, IDA, NDI linkage, prognostic models development, and prognostic model use in general medical practice. Project PERIL lays the foundation for a future implementation trial of new prognostic models informed by social circumstances and personality in general medical practice. Finally, Project PERIL provides an infrastructure for future US longevity research.